seismic assessment of masonry buildings according · pdf fileseismic assessment of masonry...
TRANSCRIPT
SEISMIC ASSESSMENT OF MASONRY BUILDINGS ACCORDING TO
EUROCODE 6 AND 8
Dimitra N. STAVRELI 1
and Stephanos E. DRITSOS 2
ABSTRACT
It is clear from several analyses that, in comparison with buildings designed according to new codes, the existing building stock has a lower seismic capacity. The problem is more serious for masonry
buildings, especially those that have been constructed in earlier years.
Despite the fact that the design code framework is clear through Eurocode 6 (2005) and Eurocode 8 (2005), the method of assessment of existing masonry structures is not obvious. It is
worth mentioning that Eurocode 6 (2005) provides only for the design of new masonry buildings and
does not consider seismic actions. This paper deals with a comparison of seismic assessment methods of masonry buildings
according to either the framework of Eurocode 6 (2005) or Eurocode 8 Part 3 (2005b) and an
approximate method proposed by the Greek Earthquake Planning and Protection Organization (EPPO,
2012). These methods are applied a selection of buildings with particular characteristics through a parametric study aiming to emphasise possible different results when assessing the seismic behaviour
of a structure. From the results of the present analysis, it is found that the extension of the
implementation of assessment procedures of Eurocode 6 (2005) to buildings that are subjected to seismic actions produces results that are very different to those that are given by the implementation of
Eurocode 8 (2005b) when the criteria for the safety verification of the building are expressed in terms
of deformations and storey drifts. Moreover, Eurocode 8 (2005b) does not consider out-of-plane
deformation of the walls while, in the framework of Eurocode 6 (2005), in-plane and out-of-plane action effects are simultaneously considered.
In the case studies investigated in the present work, Eurocode 8 (2005b) results give high failure
indices for multilevel buildings. On the other hand, Eurocode 6 (2005) results give similar failure index values from storey to storey. When comparing the Greek Earthquake Planning and Protection
method (EPPO, 2012) with Eurocodes 6 (2005) and Eurocode 8 (2005b), it appears that failure indices
from the EPPO (2012) method are always higher than unity, even in the case of simple buildings where the respective values are expected, by definition, to be much lower than unity and where the
rigorous method of assessment applied within the framework of Eurocode 8 (2005b) justified these
low failure index values. Therefore, a correction factor to multiply the resistance index of the EPPO
(2012) method is proposed in the present work. Furthermore, it is found that the results of the EPPO (2012) method are not significantly influenced by the number of storeys, which is found to be in
contrast with the results of the rigorous analytical method that was used for the same buildings.
1 PhD student, Department of Civil Engineering, University of Patras, Patras, [email protected] 2 Professor, Department of Civil Engineering, University of Patras, Patras, [email protected]
2
INTRODUCTION
Seismic assessment methods for masonry buildings according to Eurocode 6 (2005), Eurocode 8 Part
3 (2005b) and an approximate method proposed by the Greek Earthquake Planning and Protection
Organization (EPPO, 2012) are compared in this paper. It must be noted here that seismic actions are
not specifically considered by Eurocode 6 (2005), as it concerns only the design of new masonry buildings. In addition, the out-of-plane deformation of the walls is not considered by Eurocode 8
(2005b), while Eurocode 6 (2005) simultaneously considers in-plane and out-of-plane action effects.
Through a parametric study aimed to emphasise possible different results when assessing the seismic behaviour of a structure, these three methods are applied to selected buildings with particular
characteristics.
To begin with, the investigation considers a six-storey building, for which failure indices
according to Eurocode 6 (2005), Eurocode 8 (2005b) and the approximate method are computed and compared. Then, a sample of representative buildings are analysed, so the failure indices for “simple”
masonry buildings (as defined in Part 1 of Eurocode 8, 2005a) can be determined in order that some
very important conclusions can be drawn from the comparison of the three methods in such buildings. Moreover, the investigation of the influence of the number of storeys on failure indices of particular
buildings was considered crucial. For this purpose, a differentiation between a six-storey building, a
four-storey and a two-storey building is performed, so that the failure index resulting from the different methods could be compared.
MAIN CHARACTERISTICS OF MULTI-STOREY BUILDING
The six storey building that was analysed is located at the historical centre of Corfu in Greece and was
constructed during the period of the English “protection” of the Ionian Islands (1815-1864). Figure 1
presents a plan of the ground floor with the walls designated and a cross section through the six storey building.
The materials that were used for the construction of this building were typical of that time.
The walls of the basement and the ground floor storeys consist of three-leaf stone masonry in contrast to the rest of the storeys, which consist of compact brickwork of a reduced thickness. The thickness of
the walls of the basement and the ground floor are 80 cm and thereafter gradually reduce to 35 cm in
the 5th floor of the building.
Figure 1. Characteristic plan and section of the investigated six storey building
3
As far as the floors are concerned, these consist of wood and are supported by wooden beams
(of 20 by 18 cm cross section) spaced at a distance of 40 to 60 cm, on which the flooring (with a
thickness of 1.5 cm) is situated. The roof consists of a wooden truss arrangement and is covered by
tiles.
MASONRY MECHANICAL PROPERTIES
The mechanical properties of the materials were either determined from laboratory tests or by
consulting suitable references. The compressive strength for the three-leaf masonry was 10.9 MPa,
while that of the brickwork was 4.47 MPa. The shear strength in both cases was 0.07 MPa. The elastic modulus for both materials was 6516 MPa and 2682 MPa respectively and the shear modulus
was 2470 MPa and 1032 MPa respectively.
LOADING
The self weight of the two types of masonry was chosen as 21 kN/m2 for the three leaf masonry and 18
kN/m2
for the brickwork. The building has wooden floors and roof, which do not offer a significant
diaphragm action. Therefore, it was decided that only the self weight and live load loads specified by
the design code were be taken into account. As a result, dead loads of 1.30 kPa for the roof, 1.20 kPa for the floors and staircase and 0.80 kPa for internal walls were input in the analyses. The live load of
the floors and staircase was considered to be 2.20 kPa.
ANALYSIS ASSUMPTIONS
The building was modelled using the ETABS 9.5.0 (2009) computer program. The walls were
simulated using four-joint shell finite elements and were different from floor to floor according to the wall thickness. These shell elements are considered to have bending (out of plane) and membrane (in
plane) stiffness. The foundation of the building was considered as pinned. A modal analysis was
implemented. The analysis was a linear elastic multi-modal response spectrum analysis assuming a reduced stiffness for the cracked section (Eurocode 8, 2005b). The building that was analysed is
located in Corfu, so the seismic zone is II with an importance factor of class II, as it is normal building
for dwelling or offices. The soil category is C, while the behaviour factor q and the foundation factor were considered as 1.50 and 1.00 respectively. The building was checked for seismic loading
according to the Greek Seismic Code (EAK2000, 2003) with a dead and live load combination of
1.35G+1.5Q as well as the seismic load combinations of G+0.3Q±Ex±0.3Ey and G+0.3Q±0.3Ex±Ey.
FAILURE INDICES THROUGH EUROCODE 6
Through ECTools (2002) software, the cross sections of the piers and spandrels of the structure were checked for bending and shear. ECTools (2002) uses Eurocode 6 (2005) equations and failure indices
in bending and shear can be determined. Failure indices (λi) of elements are determined by dividing
action effects by capacity. Therefore, they are unitless and values greater than unity indicate that damage would occur. Obviously, the greater the number is above unity, the greater the expected
damage. According to Eurocode 6 (2005), λi in flexure with axial load is defined by the following
equation:
(1)
where:
4
Nsd is the design vertical load on a masonry section and
NRd is the design vertical load resistance of the masonry section and is given by the following
equation:
(2)
where: Φi is the capacity reduction factor allowing for the effects of slenderness and eccentricity of loading,
t is the thickness of the section and
fd is the design compressive strength of the masonry. As far as shear in Eurocode 6 (2005) is concerned, λ is defined by the following equation:
(3)
where:
Vsd is the design value of the applied shear load and
VRd is the design shear resistance and is given by the following equation:
(4)
where:
fvd is the design shear strength of the masonry and lc is the length of area under compression.
It was decided that an average failure index of the cross sections per floor for bending and shear
should be calculated. The average failure index in every floor (j), was determined from Eq. (5) and Figure 2 presents the results.
6
1
6
1
*
i
ii
ij
Ai
A
, j = 1…6 (5)
where:
λj is the failure index of the floor,
λi is the failure index of each section in bending or shear and
Α is the cross sectional area of the pier or spandrel. Figure 2 shows that the failure indices in shear almost everywhere are greater than those in
flexure. The fact that values in every storey are greater than 5.00 both in flexure and shear is also
evident. The maximum failure index value appears in the 2nd
storey of the building. The failure indices of the building (λb) in flexure and shear are the maximum failure indices that
resulted from the above analyses and are presented in Table 1.
It can be seen from Table 1 that the failure indices in flexure and shear are considerably greater
than unity and their values are very high. In addition, the failure index of the building can be considered the maximum value of flexure or shear. Therefore, λb = 11.4 is the failure index of the
building according to Eurocode 6 (2005) and occurs in the 2nd
storey.
5
Figure 2. Average failure indices according to Eurocode 6 in flexure with axial load and shear per storey
Table 1. Building failure indices in flexure and shear
λb Flexure λb Shear
9.90 11.4
FAILURE INDICES THROUGH EUROCODE 8
In this section, a computation of failure indices through the upper limit acceptance criteria of Eurocode 8 (2005b) takes place for shear and flexure with axial load respectively. The upper limit acceptance
criteria of Eurocode 8 (2005b) were computed only for in-plane deformations and are as follows:
For primary seismic walls for the "Significant Damage" performance level, the capacity of an unreinforced masonry wall controlled by flexure is expressed in terms of drift and is taken to equal
0.008Ηο/D, where Ho is the distance between the section where the flexural capacity is attained and
the point of contraflexure and D is the in-plane horizontal dimension of the wall (depth). The
respective capacity in shear is equal to the value of 0.004. For the computation of displacements and because of the fact that the displacements that result
are elastic, a behaviour factor q is used for their transformation from elastic to inelastic. Taking into
account Part 1 of Eurocode 8 (2005a), this factor was selected to be q = 1.50. Figure 3 presents the determined wall displacements.
Gr.Floor
1st floor
2nd floor
3rd floor
4th floor
5th floor
0,00 2,00 4,00 6,00 8,00 10,00 12,00
5,22
8,24
13,03
15,7
19,99
21,94
AVERAGE λj
ST
OR
EY
HE
IGH
T (
m)
Average damage index per storey Shear
Flexure with
Axial load
6
Figure 3. Wall displacements per storey
Figure 3 shows that the largest displacement occurs in wall CD, which is the longest wall. This is one of the most important influencing parameters that causes this wall to have the highest
displacement. In addition, the supporting transverse walls to this wall are not perpendicular and are at
an angle, so they do not offer a high resistance to this wall. It can also be observed that the angle of the line of displacements is normal in the first storeys, in contrast with upper storeys where the
displacement line angle notably increases, especially in walls CD and BC, which also have the feature
that their transverse walls are not perpendicular to them. This absence of normality from the 2nd
floor onwards causes higher drifts.
Having computed the displacements, the drift for each floor was determined through the
following equation:
drift =
, i = 1…6 (6)
where:
u is the in-plane displacement and
h is the floor height.
Αccording to Part 3 of Eurocode 8 (2005b), the λi failure indices are computed by dividing these drifts by their upper limit acceptance criteria for every wall of the building. The upper limit
acceptance criteria of Eurocode 8 (2005b) and the failure indices in flexure and shear for the
"Significant Damage" performance level in the walls of the building are presented in Table 2. Failure indices exceeding unity have a red font colour.
Gr. Floor
1st Floor
2nd Floor
3rd Floor
4th Floor
5th Floor
0
5
10
15
20
25
0 0,2 0,4 0,6 0,8 1
ST
OR
EY
HE
IGH
T (m
)
DISPLACEMENT (m)
Wall displacements
Wall BC
Wall CD
Wall AB
Wall AD
7
Table 2. λi failure indices through Eurocode 8 (2005b)
Wall Storey Displacement
(m) Drift
ΕC8 Flexure with Axial Load Upper Limit Acceptance
Criteria
ΕC8 Shear Upper Limit
Acceptance
Criteria
λi Flexure
with Axial
Load
λi Shear
BC Ground Floor 0.01428 0.00274 0.043 0.004 0.06 0.7
1st Floor 0.02562 0.00375 0.067 0.004 0.06 0.9
2nd Floor 0.0441 0.00587 0.106 0.004 0.06 1.5
3rd Floor 0.0651 0.00487 0.128 0.004 0.04 1.2
4th Floor 0.08127 0.00513 0.163 0.004 0.03 1.3
5th floor 0.09681 0.00503 0.179 0.004 0.03 1.3
CD Ground Floor 0.00903 0.00173 0.003 0.004 0.54 0.4
1st Floor 0.01176 0.00090 0.005 0.004 0.18 0.2
2nd Floor 0.01827 0.00207 0.008 0.004 0.26 0.5
3rd Floor 0.042 0.00551 0.010 0.004 0.57 1.4
4th Floor 0.084 0.01333 0.012 0.004 1.09 3.3
5th floor 0.11025 0.00850 0.013 0.004 0.63 2.1
ΑΒ Ground Floor 0.00126 0.00024 0.004 0.004 0.07 0.1
1st Floor 0.00231 0.00035 0.006 0.004 0.06 0.1
2nd Floor 0.00483 0.00080 0.009 0.004 0.09 0.2
3rd Floor 0.00756 0.00063 0.011 0.004 0.06 0.2
4th Floor 0.0105 0.00093 0.014 0.004 0.07 0.2
5th floor 0.01344 0.00095 0.015 0.004 0.06 0.2
AD Ground Floor 0.00252 0.00048 0.004 0.004 0.12 0.1
1st Floor 0.00504 0.00083 0.006 0.004 0.13 0.2
2nd Floor 0.00945 0.00140 0.010 0.004 0.14 0.4
3rd Floor 0.01281 0.00078 0.012 0.004 0.06 0.2
4th Floor 0.0168 0.00127 0.016 0.004 0.08 0.3
5th floor 0.01911 0.00075 0.017 0.004 0.04 0.2
Figure 4 presents the wall drifts of the previous table.
Figure 4. Wall drifts through Eurocode 8 (2005b)
Table 2 and Figure 4 show that the maximum failure index value and the maximum drift occurs
in wall CD. Some significant failure indices also appear in wall BC in contrast with walls AB and
AD, which have very low failure indices. All these findings are due to the characteristics of the walls that were described above. The most serious problem appears in the fourth floor of the building and
particularly in wall CD because of shear.
Consequently, the failure index λb for the building is the maximum λi value derived from Table 2 and, specifically, it is the value λb = 3.3, which corresponds to shear in the 4
th storey.
0
5
10
15
20
25
0,000 0,005 0,010 0,015
ST
OR
EY
HE
IGH
T (
m)
DRIFT
Wall drifts
Wall BC
Wall CD
Wall AB
Wall AD
8
COMPARISON OF EUROCODE 6 AND EUROCODE 8 FAILURE
INDICES
From the above, a comparison between the failure indices of Eurocode 6 (2005) and Eurocode 8 (2005b) is feasible. To be more exact, the maximum failure index of Eurocode 6 (2005) was 11.4. As
far as Eurocode 8 (2005b) is concerned, the maximum failure index was 3.3. Consequently, it can be
seen that the failure index of Eurocode 8 (2005b) is much smaller than that of Eurocode 6 (2005). The difference between the codes is about 245%.
This difference is due firstly to the fact that the failure indices of Eurocode 8 (2005b) concern
only in-plane drifts, which is in contrast with Eurocode 6 (2005) where the check concerns both in-plane and out-of-plane action effects. It is also clear that the failure indices of Eurocode 6 (2005) are
an average for sections, so the check is much more conservative than that of Eurocode 8 (2005b),
which focuses on the check of all the walls and their displacements per floor.
Another very important difference is that in Eurocode 6 (2005) the maximum value of failure indices appeared in 2
nd storey and there is not much of a difference from storey to storey. This is
contrary to Eurocode 8 (2005b), where the maximum value appeared in 4th storey. Consequently, it is
evident that Eurocode 8 (2005b) gives the highest failure indices in upper storeys, so these are the critical storeys for the assessment. On the other hand, Eurocode 6 (2005) gives almost the same
results as far as the storeys are concerned.
ESTIMATION OF FAILURE INDEX OF THE BUILDING ACCORDING
TO THE APPROXIMATE ASSESSMENT METHOD OF THE GREEK
EARTHQUAKE PLANNING AND PROTECTION ORGANIZATION
The Greek Earthquake Planning and Protection Organization has proposed an approximate method for
the assessment of the seismic capacity of masonry buildings (EPPO, 2012). This is based on a rational comparison between a Seismic Intensity Index (H) and a Seismic Resistance Index (R). The failure
index of a building is defined as in Εq. (7).
R
Hb (7)
The seismic intensity for the building depends on the seismic action index of the building (H1) that is defined according to the seismic zone and the influence of neighbouring buildings index (H2),
as shown in Eq. (8).
(8)
where h1 and h2 are weighting factors, which take the values of 0.75 and 0.25 respectively. The seismic resistance of a building depends on the ground floor shear resistance index R1, the
load bearing wall openings index R2, the ring beam index R3, the diaphragm index R4, the openings
near corners index R5, the masonry damage index R6, the connection between transverse walls index R7, the perimeter wall out of plane stress index R8, the ground floor plan regularity index R9 and the
height regularity index R10. The weighting factors for all these indices are presented in Table 3.
9
Table 3. Seismic resistance Indices
Index Name Weighting Factor (ri)
R1 Ground floor shear resistance index 0.20
R2 Load bearing wall openings index 0.05
R3 Ring beam index 0.15
R4 Diaphragm index 0.10
R5 Openings near corners index 0.15
R6 Masonry damage index 0.05
R7 Connection between transverse walls index 0.10
R8 Perimeter wall out of plane stress index 0.10
R9 Ground floor plan regularity index 0.05
R10 Height regularity index 0.05
Table 3 shows that the final seismic resistance depends on a number of indices and their corresponding weighting factor. Table 3 can be summarised by Eq. (9), as follows:
(9)
When computing the indices for the investigated 6-storey building, it was found that:
H1 = 2.4, H2 = 1.00 → H = 2.05 and
R1 = 0.142, R2 = 0.57, R3 = 0.50, R4 = 0.40, R5 = -1.00, R6 = 1.00, R7 = 1.00, R8 = 0.93, R9 = 0.50, R10 = 1.00 → R = 0.342.
Consequently, the value of the failure index that results through this method is: λb = Η/R = 6.00
COMPARISON OF THE EPPO METHOD FAILURE INDEX WITH THE
FAILURE INDEX OF EUROCODES 6 AND 8
According to the values of failure indices that resulted through Eurocode 6 (2005), Eurocode 8
(2005b) and the approximate seismic method (EPPO, 2012), an improvement of the failure index of
the EPPO method λb is proposed by introducing a correction factor “β” in the form of R
Hb
*'
.
When taking into account that λb values of 11.4 and 3.33 have been found according to
Eurocode 6 (2005) and Eurocode 8(2005b) respectively, the correction factor β is equal to 0.53 and
1.79 respectively.
FAILURE INDICES FOR “SIMPLE” MASONRY BUILDINGS
With the aim to expand the sample of representative buildings that are analysed, the failure indices for
“simple” masonry buildings have been computed. According to Part 1 of Eurocode 8 (2005a),
“simple” buildings are those which belong to importance classes I and II and an explicit safety
verification is not mandatory for them. Consequently, according to certain rules, these buildings are considered safe without checking the integrity of them by any specific seismic analysis. Therefore, it
could be expected that seismic assessment of capacity of those buildings would result in failure indices
somewhat lower than unity. It is obvious that the configuration parameters of a building are too numerous. Therefore, it was
initially decided to consider concrete slabs in order to ensure an effective diaphragm action.
In order to broaden the investigation of the present work, it was decided to investigate three buildings with a cross sectional area greater than the minimum (Fig. 5 left) and two buildings with
different minimum areas of cross sections per direction as a percentage of the area of the above floors
(Fig. 5 right).
Figure 5 presents typical plans and includes the dimensions of the piers and spandrels.
10
Figure 5. “Simple” building plans A, B, C (left) and D, E (right)
The main parameters of the buildings that were investigated are given in Table 4 along with the
average failure indices of Eurocode 6 (2005), the maximum values of the failure indices of Eurocode 8 (2005a) and the failure indices of the EPPO method (EPPO, 2012).
Table 4. Main characteristics and failure indices of the investigated “simple” buildings
Building Seismic
Zone
Number of
Storeys
Masonry Thicknes
s (m)
Minimum Pier Area
(%)
λb
EC6 Flexure
EC6 Shear
EC8 Flexure
EC8 Shear
EPPO
Α I 1 0.25 7.58 1.82 1.95 0.24 0.20 1.57
Β I 2 0.25 7.58 2.14 1.67 0.29 0.25 1.65
C II 1 0.30 9.1 3.2 1.51 0.29 0.25 2.21
D II 1 0.30 6.0 4.16 1.96 0.38 0.32 2.40
Ε I 2 0.25 5.0 2.78 2.17 0.38 0.33 1.70
Table 4 shows that the factors that influence the failure indices of simple buildings are the
number of storeys, the cross sectional area of the piers as percentage of the area of above floors and the seismic zone.
According to the above and having chosen buildings D and E that have a minimum area of piers
for a particular seismic zone, a correction of the EPPO (2012) λb failure indices can be performed so
that, according to Eurocode 6 (2005) and Eurocode 8 (2005b) respectively, the maximum values of the EPPO (2012) λb will define if a building can be determined as “simple” or not.
Following the same method, the maximum factors β derived from flexure or shear for buildings
B and C can be determined, as presented in Table 5. The aim here is to compare the failure indices of buildings that have the same parameters with the exception of the percentage of pier area.
Consequently, possible β values for the EPPO (2012) λb factors can be determined, as shown in Table
5.
Table 5. β correction factor values
Building Seismic
Zone
Number of
Storeys
Masonry
thickness (m)
Minimum
pier area (%)
β
EC6 EC8
D II 1 0.30 6.0 1.22 7.69
C II 1 0.30 9.1 1.47 9.09
Ε I 2 0.25 5.0 0.78 5.26
Β I 2 0.25 7.58 0.99 6.67
From Table 5, it is evident that the factors obtained from Eurocode 6 (2005) and Eurocode 8 (2005) are as expected and are in line with the six-storey building results, that is, there are much lower
values when Eurocode 6 (2005) is compared to Eurocode 8 (2005b). Therefore, since Eurocode 6
(2005) values are far more conservative than those of Eurocode 8 (2005b), the H/R factors do not exceed unity. Considering Eurocode 6 (2005), the results could be characterised as being fairly close
11
to the H/R values and, thus, the obtained correction factors are close to unity. It can also be observed
that according to Eurocode 8 (2005b) values (which indicate that “simple masonry buildings” are
structurally more than adequate), H/R failure indices are too conservative for one or two storey
buildings, as the derived correction factors are much higher than unity. Through the inadequacies of Eurocode 8 (2005b), it can be concluded that H/R indices should be suitably modified in order to also
assume values less than unity, as in the case of “simple masonry buildings”.
INFLUENCE OF NUMBER OF STOREYS ON THE FAILURE INDICES
OF THE BUILDING
A comparison of six-storey buildings to four-storey and two-storey buildings is necessary, so that the failure λ of Eurocode 8 (2005b) and λb of EPPO (2012) can be computed. Taking also into account the
respective results in “simple” one-storey and two-storey buildings, a very safe conclusion as far as the
influence of the number of storeys on the failure indices will be feasible.
The second level pre-earthquake assessment method is particularly conservative in cases of one or two storey buildings, as the relevant inadequacies are greater than unity while, according to
Eurocode 8 (2005b) deficiency results, it is highly likely that there are adequacies in such buildings.
The latter is evident not only from the “simple masonry buildings” inadequacies, but also from the six storey building variations (4 storey and 2 storey), where the difference when compared to the six
storey deficiencies is significant. For this purpose, the failure indices of five buildings using Eurocode
8 and the approximate assessment method (EPPO, 2012) are presented in Table 6 and Figure 6.
Table 6. Failure indices of "simple" buildings and 2 storey, 4 storey and 6 storey buildings
Building λb EC8 (Flexure & Axial
Load) λb EC8 (Shear)
λb EPPO
“Simple” 1 storey (A) 0.24 0.20 1.57
“Simple” 2 storey (B) 0.29 0.25 1.65
2 storey (C) 0.48 0.60 5.20
4 storey (D) 0.92 1.47 5.70
6 storey (E) 1.09 3.36 6.00
Figure 6. Failure indices in “simple” 1 storey, “simple” 2 storey, 2 storey, 4 storey and 6 storey buildings
From the above table and figure, it can be seen that the EPPO (2012) λb indices gradually icrease when adding storeys until a maximum value, which is 6.00 for the 6 storey building. This is in
contrast with the results of the failure indices of Eurocode 6 (2005), which steadily increase as the
number of storeys increase. On the contrary, the failure indices calculated through Eurocode 8
0
1
2
3
4
5
6
7
A B C D E
DA
MA
GE
IN
DIC
ES
BUILDING
Damage indices of buildings
λb (Flexure)
λb (Shear)
λb EPPO
12
(2005b) rapidly increase when storeys are added. From the above, it can be concluded that the R1
index (shear resistance index) of the approximate assessment method (EPPO, 2012) should be revised
when taken into account the number of storeys and, more importantly, the influence of the number of
the storeys should be reflected in the other seismic resistance indices of the building.
CONCLUSIONS
When assessing the seismic capacity of masonry buildings in earthquake regions, Eurocode 6 (2005)
appears to be much more conservative when compared to Eurocode 8 (2005b), especially for one or
two storey buildings and “simple masonry buildings” as defined in Part 1 of Eurocode 8 (2005a). Within the framework of Eurocode 6 (2005), action effects are compared with relevant
resistances in terms of forces, while in the framework of Eurocode 8(2005b) they are compared in
terms of deformations (drifts). Moreover, Eurocode 8 (2005b) does not consider out-of-plane wall
deformation while the framework of Eurocode 6 (2005) considers both in-plane and out-of plane action effects simultaneously.
In the case studies investigated in the present work, it was found that Eurocode 8 (2005b) gives
the highest failure indices for higher level buildings. On the other hand, Eurocode 6 (2005) results in similar failure index values from storey to storey.
When comparing the Greek Earthquake Planning and Protection method (EPPO, 2012) with
Eurocode 6 (2005) and Eurocode 8 (2005b), it appears that failure index values for the EPPO (2012) method are always higher than unity, even in cases of "simple" buildings where respective values, by
definition, would be expected to be much lower than unity and the rigorous method of assessment
applied in the framework of Eurocode 8 (2005b) justifies these low failure index values. Therefore, a
correction factor multiplying the resistance index of the EPPO (2012) method has been proposed by the present work.
Finally, it was found that the results of the EPPO (2012) method are not much influenced by the
number of storeys, which was found to be in contrast with the results of the rigorous analytical method that was used for the same buildings. As a result and through the present investigation, it can be
concluded that it is necessary that the number of storeys should have greater influence when
determining the seismic resistance of a building.
AKNOWLEDGEMENTS
The authors would like to acknowledge the assistance of 3π - Penelis Software Ltd and Computers & Structures Inc. by providing free licences for their software. Many thanks are expressed to Dr V. J.
Moseley for his assistance during the preparation of this paper.
REFERENCES
EAK2000 (2003) “Greek code for seismic resistant structures”, Earthquake Planning and Protection
Organization, Greek Ministry for Environmental Planning and Public Works, Athens, Greece (in Greek)
ECTools (2002) 3π - Penelis Software Ltd, http://www.ectools.eu/company
EPPO (2012) “Second level pre-earthquake assessment for masonry buildings”, Earthquake Planning and Protection Organization, Greek Ministry for Environmental Planning and Public Works, Athens, Greece
(draft in Greek)
ETABS (2009) ETABS users manual, Computers & Structures Inc., Berkeley, California
Eurocode 6 (2005) “Design of masonry structures-Part 1-1: General rules for reinforced and unreinforced
masonry structures”, ENV 1996-1-1: 1995
Eurocode 8 (2005a) “Design of structures for earthquake resistance”, Part 1 “General rules, Seismic actions and
rules for buildings”, EN 1998-1
Eurocode 8 (2005b) “Design of structures for earthquake resistance”, Part 3 “Assessment and retrofitting of
buildings” EN 1998-3: 1995